Exploring the tidal effect of urban business district with large-scale human mobility data

doi:10.1007/s11704-022-1623-6

Front. Comput. Sci.

2023, Vol. 17

Issue (3) : 173319 https://doi.org/10.1007/s11704-022-1623-6

RESEARCH ARTICLE

Exploring the tidal effect of urban business district with large-scale human mobility data

Hongting NIU¹(

), Ying SUN^2,³, Hengshu ZHU³, Cong GENG¹, Jiuchun YANG⁴, Hui XIONG⁵, Bo LANG¹

¹. School of Computer Science and Engineering, Beihang University, Beijing 100191, China
². Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing Technology, CAS, Beijing 100086, China
³. Baidu Talent Intelligence Center, Baidu Inc., Beijing 100085, China
⁴. Business School, Imperial College London, London SW72AZ, UK
⁵. Artificial Intelligence Thrust, The Hong Kong University of Science and Technology, Guangzhou 510030, China

Download: PDF(18787 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Business districts are urban areas that have various functions for gathering people, such as work, consumption, leisure and entertainment. Due to the dynamic nature of business activities, there exists significant tidal effect on the boundary and functionality of business districts. Indeed, effectively analyzing the tidal patterns of business districts can benefit the economic and social development of a city. However, with the implicit and complex nature of business district evolution, it is non-trivial for existing works to support the fine-grained and timely analysis on the tidal effect of business districts. To this end, we propose a data-driven and multi-dimensional framework for dynamic business district analysis. Specifically, we use the large-scale human trajectory data in urban areas to dynamically detect and forecast the boundary changes of business districts in different time periods. Then, we detect and forecast the functional changes in business districts. Experimental results on real-world trajectory data clearly demonstrate the effectiveness of our framework on detecting and predicting the boundary and functionality change of business districts. Moreover, the analysis on practical business districts shows that our method can discover meaningful patterns and provide interesting insights into the dynamics of business districts. For example, the major functions of business districts will significantly change in different time periods in a day and the rate and magnitude of boundaries varies with the functional distribution of business districts.

Keywords business district trajectory functionality detection tidal effect boundary detection visiting score

Corresponding Author(s): Hongting NIU

About author:

Tongcan Cui and Yizhe Hou contributed equally to this work.

Just Accepted Date: 15 March 2022 Issue Date: 08 September 2022

Cite this article:

Hongting NIU,Ying SUN,Hengshu ZHU, et al. Exploring the tidal effect of urban business district with large-scale human mobility data[J]. Front. Comput. Sci., 2023, 17(3): 173319.

URL:

https://academic.hep.com.cn/fcs/EN/10.1007/s11704-022-1623-6
https://academic.hep.com.cn/fcs/EN/Y2023/V17/I3/173319

Fig.1 Radiation range and distribution of the business district in Chengdu, China. (a) Radiation map is generated by taking the center of Chengdu’s main business district as the pick-up point or drop-off point, and taking the line connecting the order pick-up point and the drop-off point as the radial line segment [3]; (b) distribution of some business districts and their approximate radiation area [4]

Tab.1 Number of POI in each category

Fig.2 Statistics of boundary score. (a) Location of main business districts; (b) total number of grids in each level of boundary score; (c) number of grids covered by the business district in each boundary level

Tab.2 Name and location coordinates of the main business districts based on boundary data in Chengdu, China

Notation	Description
$N ? M$	Number of city grids
$ξ$	Time slices in a whole day
$T d$	Drop-off flow
$T p$	Pick-up flow
$T d d$	Drop-off flow in the day
$T p d$	Pick-up flow in the day
$T d n$	Drop-off flow in the night
$T p n$	Pick-up flow in the night
$P n$	POI number
$P c n$	POI category number
$P c i$	POI category name (20)
$P n c i$	POI number of categories (20)
$G l o c$	Location of grid (ID)
$G L$	The most POI class
$G s$	The most POI subclass
$G R$	Weather remote in the city
$G m$	Weather long/short mileage of order
$V m l$	Number of long mileage orders
$V m s$	Number of short mileage orders
$S b o u n d a r y$	The grade of boundary score
$S v i s i t i n g$	The grade of visiting score

Tab.3 Notation for features

Fig.3 Framework of boundary and functionality pattern detection

Fig.4 Schematic diagram for HITS algorithm. (a) Hub value; (b) authority value

Notation	Description	Dimension
$G l o c$	Location of grid	2
$T p$	Pick-up flow	1
$T d$	dorp-off flow	1
$[T p 1, T p 2, . . .]$	Pick-up flow	8
	in surrounding 8 grids
$[T d 1, T d 2, . . .]$	Drop-off flow	8
	in surrounding 8 grids
$P n c i$	POI number of categories	20

Tab.4 Construction of a totally

40

-dimensional feature vector construction for boundary pattern detection

Tab.5 Precision of ranking list based on HITS in different data set size

Tab.6 Estimation of boundary prediction with supervised method

Fig.5 Results of classification evaluation at the hierarchical boundary. (a) Random forest; (b) decision tree; (c) naive bayes; (d) SVM

Fig.6 Classification result of boundary score. All grids are colored with several grades. Set the score from high to low to green, blue, yellow and orange. The closer to the center of business districts, the higher of boundary score. The further away from the business district center, the lower of of boundary score. This is in line with our general cognitive process

Fig.7 Statistics of the number of grids inside the boundary in the tidal cycle of a day. Y-axis is counted by the number of boundary score

S b o u n d a r y ≥ 1

, X-axis is

m

-time slices. (a) Changing trend on 24-time slices; (b) changing trend on 144-time slices

Fig.8 Boundary evolution with time periods. The color bar is placed on the right side of each row of graphs. The brightest color is yellow with the highest score, the least bright color is green with the lowest score. The brighter the color, the higher the boundary score of the business district. (a) 0:00; (b) 4:00; (c) 8:00; (d) 12:00; (e) 16:00; (f) 20:00

Tab.7 Performance comparison of each method on

144

-time slices and

24

-time slices

Index	(a) Original			(b) Optimal
Index	$P o r i$	$R o r i$	$F o r i$	$P o p t$	$R o p t$	$F o p t$
Total	0.8640	0.9761	0.9166	0.8826	0.9734	0.9258
Weekdays	0.8856	0.9756	0.9284	0.9011	0.9729	0.9356
Weekends	0.8448	0.9671	0.9018	0.8603	0.9685	0.9112

Tab.8 Performance comparison before and after the optimization of boundary threshold based on LSTM

Fig.9 Statistics about the threshold of boundary score. (a) Performance of the model at different moments; (b) boundary value when the F-Score is maximum at different moments

Tab.9 Evaluation of the precision index of functionality prediction

Fig.10 Results of ablation evaluation in different business districts. V_o, V_a and V_c separately represent the vector of total features, self-related features, and radiation-related features

Fig.11 Functionality evolution in one day of the three typical business districts. (a) Jianshe Road; (b) Luomashi; (c) Shuangnan

Fig.12 Boundary and functionality changing case of the typical business districts. The area marked with an orange box is the area where the boundary has changed. (a) Case for the boundary change of the Wuhouci business district, 2 reduced grids’ function is Home; (b) case for the boundary change of the Chunxi Road business district, 4 reduced grids’ function is Food

1	F, Wang X, Gao Z Xu . Identification and classification of urban commercial districts at block scale. Geographical Research, 2015, 34( 6): 1125– 1134
2	J, Xiao Y, Shen J, Ge R, Tateishi C, Tang Y, Liang Z Huang . Evaluating urban expansion and land use change in Shijiazhuang, China, by using GIS and remote sensing. Landscape and Urban Planning, 2006, 75( 1−2): 69– 80
3	D D Institute. 2018 China Urban Business Circle Travel and Consumption Analysis Report. Business district radiation map drawn by didi travel big data. See 199it website, 2018
4	Star News Red. Schematic diagram of the distribution of Chengdu’s business districts and the density levels of business districts. See Sohu website, 2020
5	J M Kleinberg . Authoritative sources in a hyperlinked environment. Journal of ACM, 1999, 46( 5): 604– 632
6	X, Shi Z, Chen H, Wang D Y, Yeung W K, Wong W C Woo. Convolutional LSTM network: a machine learning approach for precipitation nowcasting. In: Proceedings of the 28th International Conference on Neural Information Processing Systems. 2015, 802– 810
7	N J, Yuan Y, Zheng X, Xie Y, Wang K, Zheng H Xiong . Discovering urban functional zones using latent activity trajectories. IEEE Transactions on Knowledge and Data Engineering, 2015, 27( 3): 712– 725
8	A, Graves N Jaitly. Towards end-to-end speech recognition with recurrent neural networks. In: Proceedings of the 31st International Conference on International Conference on Machine Learning. 2014, II-1764− II-1772
9	H, Niu H, Zhu Y, Sun X, Lu J, Sun Z, Zhao H, Xiong B Lang . Exploring the risky travel area and behavior of car-hailing service. ACM Transactions on Intelligent Systems and Technology, 2022, 13( 1): 9
10	G, Ke Q, Meng T, Finley T, Wang W, Chen W, Ma Q, Ye T Y Liu. LightGBM: a highly efficient gradient boosting decision tree. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, 3149– 3157
11	A, Pérez P, Larrañaga I Inza . Supervised classification with conditional Gaussian networks: increasing the structure complexity from naive Bayes. International Journal of Approximate Reasoning, 2006, 43( 1): 1– 25
12	M, Dumont R, Marée L, Wehenkel P Geurts. Fast multi-class image annotation with random subwindows and multiple output randomized trees. In: Proceedings of the 4th International Conference on Computer Vision Theory and Applications. 2009, 196– 203
13	L Breiman . Random forests. Machine Learning, 2001, 45( 1): 5– 32
14	J A K, Suykens J Vandewalle . Least squares support vector machine classifiers. Neural Processing Letters, 1999, 9( 3): 293– 300
15	M, Schmidt Roux N, Le F Bach. Minimizing finite sums with the stochastic average gradient. Mathematical Programming, 2017, 162( 1– 2): 1– 2
16	F, Pedregosa G, Varoquaux A, Gramfort V, Michel B, Thirion O, Grisel M, Blondel P, Prettenhofer R, Weiss V, Dubourg J, Vanderplas A, Passos D, Cournapeau M, Brucher M, Perrot É Duchesnay . Scikit-learn: machine learning in python. The Journal of Machine Learning Research, 2011, 12: 2825– 2830
17	W, Yu T, Ai S Shao . The analysis and delimitation of Central Business District using network kernel density estimation. Journal of Transport Geography, 2015, 45: 32– 47
18	D L Huff . A probabilistic analysis of shopping center trade areas. Land Economics, 1963, 39( 1): 81– 90
19	B, Hao S, Dong Y C, Hu X, Liu Y J, Gao Y D Zhang . Urban business zones delimitation method based on the fusion of multidimensional characteristics. Geography and Geo-Information Science, 2017, 33( 5): 56– 62
20	G, Qi X, Li S, Li G, Pan Z, Wang D Zhang. Measuring social functions of city regions from large-scale taxi behaviors. In: Proceedings of 2011 IEEE International Conference on Pervasive Computing and Communications Workshops (PERCOM Workshops). 2011, 384– 388
21	J, Yuan Y, Zheng X Xie. Discovering regions of different functions in a city using human mobility and POIs. In: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012, 186– 194
22	H, Dong M, Wu X, Ding L, Chu L, Jia Y, Qin X Zhou . Traffic zone division based on big data from mobile phone base stations. Transportation Research Part C: Emerging Technologies, 2015, 58: 278– 291
23	Y, Liu F, Wang Y, Xiao S Gao . Urban land uses and traffic ’source-sink areas’: evidence from GPS-enabled taxi data in Shanghai. Landscape and Urban Planning, 2012, 106( 1): 73– 87
24	G, Pan G, Qi Z, Wu D, Zhang S Li . Land-use classification using taxi GPS traces. IEEE Transactions on Intelligent Transportation Systems, 2013, 14( 1): 113– 123
25	P, Zhang Z, Bao Y, Li G, Li Y, Zhang Z Peng. Trajectory-driven influential billboard placement. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2018, 2748– 2757
26	Y, Sun H, Zhu F, Zhuang J, Gu Q He. Exploring the urban region-of-interest through the analysis of online map search queries. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018, 2269– 2278
27	S, Wang Z, Bao J S, Culpepper G Cong . A survey on trajectory data management, analytics, and learning. ACM Computing Surveys, 2022, 54( 2): 39
28	D, Wang T, Miwa T Morikawa . Big trajectory data mining: a survey of methods, applications, and services. Sensors, 2020, 20( 16): 4571
29	M, Lu Z, Wang X Yuan. TrajRank: exploring travel behaviour on a route by trajectory ranking. In: Proceedings of 2015 IEEE Pacific Visualization Symposium (PacificVis). 2015, 311– 318
30	Y, Zheng G, Zhao J Liu. A novel grid based k-means cluster method for traffic zone division. In: Proceedings of the 2nd International Conference on Cloud Computing and Big Data. 2015, 165– 178
31	G, Sun B, Chang L, Zhu H, Wu K, Zheng R Liang . TZVis: visual analysis of bicycle data for traffic zone division. Journal of Visualization, 2019, 22( 6): 1193– 1208
32	Y, Miyagi M, Onishi C, Watanabe T, Itoh M Takatsuka . Classification and visualization for symbolic people flow data. Journal of Visual Languages & Computing, 2017, 43: 91– 102
33	H, Ren S, Ruan Y, Li J, Bao C, Meng R, Li Y Zheng. MtrajRec: map-constrained trajectory recovery via Seq2Seq multi-task learning. In: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2021, 1410– 1419
34	P, Han J, Wang D, Yao S, Shang X Zhang. A graph-based approach for trajectory similarity computation in spatial networks. In: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 2021, 556– 564
35	H, Wu L, Liu Y, Yu Z, Peng H, Jiao Q Niu . An agent-based model simulation of human mobility based on mobile phone data: how commuting relates to congestion. ISPRS International Journal of Geo-Information, 2019, 8( 7): 313
36	X, Chen J, Wang K Xie. TrafficStream: a streaming traffic flow forecasting framework based on graph neural networks and continual learning. In: Proceedings of the 30th International Joint Conference on Artificial Intelligence. 2021, 3620– 3626
37	Z, Fang Q, Long G, Song K Xie. Spatial-temporal graph ODE networks for traffic flow forecasting. In: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 2021, 364– 373
38	H, Wan Y, Lin S, Guo Y Lin. Pre-training time-aware location embeddings from spatial-temporal trajectories. IEEE Transactions on Knowledge and Data Engineering, 2021, DOI: https://doi.org/10.1109/TKDE.2021.3057875
39	C, Cao M Li. Generating mobility trajectories with retained data utility. In: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 2021, 2610– 2620
40	C Y, Chow M F Mokbel . Trajectory privacy in location-based services and data publication. ACM SIGKDD Explorations Newsletter, 2011, 13( 1): 19– 29
41	Y, Kim J, Han C Yuan. TOPTRAC: topical trajectory pattern mining. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2015, 587– 596
42	D W, Choi J, Pei T Heinis . Efficient mining of regional movement patterns in semantic trajectories. Proceedings of the VLDB Endowment, 2017, 10( 13): 2073– 2084

[1]

FCS-21623-OF-HN_suppl_1

Download

[1]	Xiao PAN, Lei WU, Fenjie LONG, Ang MA. Exploiting user behavior learning for personalized trajectory recommendations[J]. Front. Comput. Sci., 2022, 16(3): 163610-.
[2]	Hao LIN, Guannan LIU, Fengzhi LI, Yuan ZUO. Where to go? Predicting next location in IoT environment[J]. Front. Comput. Sci., 2021, 15(1): 151306-.
[3]	Satoshi MIYAZAWA, Xuan SONG, Tianqi XIA, Ryosuke SHIBASAKI, Hodaka KANEDA. Integrating GPS trajectory and topics from Twitter stream for human mobility estimation[J]. Front. Comput. Sci., 2019, 13(3): 460-470.
[4]	Zhigang ZHANG, Xiaodong QI, Yilin WANG, Cheqing JIN, Jiali MAO, Aoying ZHOU. Distributed top-k similarity query on big trajectory streams[J]. Front. Comput. Sci., 2019, 13(3): 647-664.
[5]	Jiali MAO, Qiuge SONG, Cheqing JIN, Zhigang ZHANG, Aoying ZHOU. Online clustering of streaming trajectories[J]. Front. Comput. Sci., 2018, 12(2): 245-263.
[6]	Chengliang WANG,Yayun PENG,Debraj DE,Wen-Zhan SONG. DPHK: real-time distributed predicted data collecting based on activity pattern knowledge mined from trajectories in smart environments[J]. Front. Comput. Sci., 2016, 10(6): 1000-1011.
[7]	Jia WEN, Chao LI, Zhang XIONG. Behavior pattern extraction by trajectory analysis[J]. Front Comput Sci Chin, 2011, 5(1): 37-44.

Viewed

Full text

Abstract

Cited

Shared

Discussed